Transient behavior following IP 3 stimulus

Calcium release from the stores requires transient stimulation of IP3R by IP3. Literature measurements of IP3 time courses in human platelets stimulated by strong doses of agonists such as thrombin and ADP show [IP3] rising sharply by several fold and peaking within 2.5 – 30 s following agonist delivery [132, 144, 145]. Furthermore, [IP3] and [Ca2+]cyt return to nearly resting levels 30 – 60 s following agonist stimulation [130] as IP3 is degraded by phosphatases and Ca2+ is pumped back into the DTS. Figure 3.3 shows the response of a subset of the fully filtered population of resting states to an IP3 stimulus constructed to mimic these measured behaviors (see Section 2.5).

Figure 3.3 Representative model responses to IP3 stimulus.

15 resting states satisfying all filtering criteria were stimulated with 5.5 fold IP3 dose starting at t = 20 s. The IP3 doses are illustrated in the inset in the upper right panel. (B) Calcium traces for DTS and cytosolic Ca2+ of a representative configuration taken from this population to the six IP3 doses shown in the inset in the upper right panel. The colors of the curves correspond to the IP3 dose. Left column: [Ca2+]ex = 1.2 mM; right column: [Ca2+]ex = 1 pM.

[Ca2+]cyt generally peaks within 15 – 20 s of IP3 application (the [IP3] time courses which are shown in the inset peak ~3 s following application), and reaches a higher peak and remains more sustained in the presence of extracellular calcium. [Ca2+]dts partially refills in simulations run with Caex and shows no refilling in calcium-free media. [Ca2+]cyt peaks earlier in simulations without Caex; this is due to SOCE continuing to be partially active

as stores refill rather than there being a time delay in ISOC activation following store depletion (see Figure 3.6 A).

Figure 3.4 B shows the calcium responses of a representative configuration from the stable population to a series of increasing [IP3] stimulations shown in Figure 3.4 A. The 1.5x and 2x stimulations were too weak to elicit substantial SOCE; at higher IP3 doses, stores become depleted enough that SOCE manifests itself via elevated [Ca2+]cyt and substantial store refilling. The model predicts faster time to peak [Ca2+]cyt at stronger IP3 doses in –Caex simulations; in contrast, in +Caex simulations peak [Ca2+]cyt is higher and delayed until stores refill to sufficiently inactivate SOCE (also see Figure 3.6).

Figure 3.4 Response of population stable in EDTA to a dose of IP3.

(A) The population of states satisfying the stability criterion (n = 18,932) were stimulated with a dose of IP3 which is summarized as a percentage increase over baseline. The IP3 stimulus begins at 10 s. (B) Calcium traces for DTS and cytosolic Ca2+ of a representative configuration taken from this population to the six IP3 doses presented in panel A. The colors of the curves correspond to the IP3 dose. Left column: [Ca2+]ex = 1.2 mM; right column: [Ca2+]ex = 1 pM. (C) Distributions of peak [Ca2+]cyt for each IP3 dose. The black vertical dotted line indicates mean resting [Ca2+]cyt and the red dotted line indicates our cutoff for a configuration to be considered “responsive” to stimulus. The number and percentage of configurations satisfying this criterion are indicated in red in the upper right of each histogram. Increasing the IP3 dose makes more states satisfy this criterion; in addition peak [Ca2+]cyt largely saturates beyond 5x IP3.

Figure 3.4 C shows histograms of peak [Ca2+]cyt of the entire set of stable configurations (n = 18,932) following application of each dose of IP3 in panel A. Stronger doses of IP3 predictably result in higher average [Ca2+]cyt; this effect saturates past 5x IP3.

Figure 3.5 compares the model’s [Ca2+]cyt response to the 10x IP3 stimulation (left) with experimental time course data (right) from human PRP fluorescently labeled with fura-2 and treated with 40 M ADP at t = 20s [150]. The simulations generally agree very well with the experimental data; there is a small time delay of ~20s in peak [Ca2+]cyt in the +Caex simulation compared to experiment; this time delay is reproducible using other model initial conditions (data not shown).

Figure 3.5 Comparison of model output with experimentally measured calcium

transient.

The 10x IP3 simulated [Ca2+]cyt time course from Figure 3.3 B (left) is compared with calibrated fura-2 time course data in human platelets treated with 40 M ADP at t = 20 s (right) [150].

= 10s. The corresponding [Ca2+]cyt and [Ca2+]dts traces are shown in Figure 3.6 B with peak [Ca2+]dts and [Ca2+]cyt times indicated by the bolded solid and dashed lines, respectively. SOCE flux peaks at the time [Ca2+]dts reaches its minimum. This implies there is no delay in SOCE activation with respect to [Ca2+]dts. As stores do not refill immediately, there is a slow inactivation of SOCE current which continues to drive [Ca2+]cyt higher even after the stores have finished emptying; [Ca2+]cyt does not peak until the SOCE flux is nearly overtaken by the PMCA flux out of the cell.

Figure 3.6 Time delay in +Caex simulations is due to slow SOCE inactivation.

(A) This panel plots Ca2+ flux across the PM, the blue curve being Ca2+ flux out of the cell via PMCA, and the red curve Ca2+ flux into the cell via open SOCs. ISOC peaks at about the time when [Ca2+]dts reaches its minimum value (peak time indicated by a vertical solid black line). [Ca2+]cyt does not peak (peak time indicated by a vertical dashed black line) until the extra Ca2+ influx into the cytosol is reduced to near zero by PMCA. (B) Calcium traces of [Ca2+]cyt and [Ca2+]dts corresponding to the simulation shown in A; each are normalized by their maximum value. The bolded vertical lines are the same as in panel A.

In Figure 3.7, model configurations in the stable in EDTA population were subjected to a 10x IP3 dose at t = 0 s with [Ca2+]ex set to 1.2 mM. Panel A shows no correlation between the time at which [Ca2+]cyt reaches its peak and fractional store

refilling. Panel B shows that the peak [Ca2+]cyt time decreases as the SERCA to IP3R ratio increases. As SERCA and IP3R act in direct opposition to one another, configurations with a higher SERCA / IP3R ratio are able to refill stores more quickly.

Figure 3.7 Rate of store refilling is an important determinant of peak [Ca2+]cyt

time in +Caex simulations.

(A) Scatter plot of the time at which [Ca2+]cyt peaks versus fractional amount of store refilling reveals no correlation between the metrics. Fractional refilling is calculated as the final value of [Ca2+]dts in a simulation divided by the initial drop following IP3 stimulus. (B) Scatter plot of the time at which [Ca2+]cyt peaks versus the SERCA / IP3R ratio.